Is the Trump Surge in Polls Here to Stay?
By Luis H. Ball Jr.
The story four days out from election day appears to be that of a late Trump surge on the heels of the renewed Hillary Clinton email scandal. And yet, does the comparison between Donald J. Trump’s late surge and Mitt Romney’s slump in 2012 really hold up?
— Luis H. Ball Jr. (@luqven) November 3, 2016
This election cycle, there are 10 percent more undecided voters than in 2012, more early voters, more latino voters and less black american turnout. So what stake, if any, can we really put into the comparison? Most pollsters agree that the key to this election is likely going to be found in the state of Nevada.
Nevada’s gaming industry has historically made it difficult for voters in the state to line up and cast their ballots on election day. This contributes to their unusually high, approximately 70%, early voting participation. The high number of early voters makes Nevada the ideal election predictor on several fronts.
The bad news? The race in Nevada is too close to call.
For starters, the poll for the state from FiveThirtyEight have the candidates in a virtual dead heat. Some pollsters argue the Johnson estimates are high given his recent implosion-prone campaign, but then again, by the same argument so might Hillary’s. The tight overall polling data is by no means the only way to predict the state’s outcome however, as some pollsters claim defeat is all but certain for the Republican presidential candidate for one key reason: Latino turnout.
- Read more: US Expats in Mexico Fear Retaliation Should Trump Become President
- Read more: As US Elections Intensify, How Will Hillary Handle Her Husband’s Past?
In Nevada, a state that between 1968 and 2004 went Democratic just twice, Latino turnout was key to delivering Obama’s victories in ’08 and ’12. This trend looks likely to repeat itself as data suggests that Democrats have a 13.6 point — or 48,000 vote — lead over Republicans in early voting. Exactly how much of that turnout is actually Latino, however, is hard to say.
Polling in Nevada, despite the early voting data, is notoriously bad (FiveThirtyEight). The population is highly transient and since 68 percent of votes come from Clark City County, getting the county wrong can ruin entire polls, as Mr. Mellman points out.
Despite the fact that Nevada’s population is 36 percent latino, it’s hard to know exactly how much non-college-educated voters, which are predominantly pro-Trump, might counteract them. If the race there was less close then one could confidently say Clinton clinched Nevada, but as it stands, with her emails in-tow, the state is as much a tossup as Florida, Ohio, or Iowa.
Moving forward, Nevada will prove to be an effective election outcome predictor. High latino turnout there could spell doom for the Trump campaign elsewhere, especially Florida. But four days away from election day, the state is simply too close to call. What is indisputable at this point, however, is that unlike in 2012, the blue-firewall of traditional Democratic states has been eroded to be “more of a rusting, chain-link fence.”
Win Probability Changes In Last Two Weeks
This could simply be explained away by saying that half of the entire electorate of 2012, whites without a college degree, overwhelmingly favor Trump. With 44 percent of the electorate in the bag, can Hillary really ride the Latino vote all the way form Nevada, to Florida, and then the White House?
For better or worse, the momentum lies with the Republican nominee.